3 mini AI apps in 16 days
I built 3 mini AI apps in the last 16 days using AI. This would definitely not have been possible without AI in the past. It is so satisfying to go from idea to live prototype in a couple of hours.
I try to limit the development time per project to 4 hours, while other things like planning the project, writing the content, social posts, etc. take another 4 hours or so.
Give it a try and let me know what you think.
http://leancanvasai.com
http://officehoursai.com
http://getfeedbackai.com
I’ll share some lessons early next week on the process, insights, what's next, etc.
Trending on Indie Hackers
Maybe a dumb question - does anyone know how much it costs to host these things? I see all these AI apps popping up but with no revenue model. I'm curious if they are just paying out of pocket for server hosting + OpenAI API fees?
Hi @mitstudent2024 I wrote a longer post here: https://aisuperfounder.substack.com/p/3-mini-ai-apps-launched-in-16-days
Great question! I'll share more exact breakdowns/details in my write-up later this week. I am running all my mini projects on the same server and database to save time.
Short version:
Assuming we did not receive any free credits, it would cost about $50/month to run these little experiments with my chosen stack.
Curious, why not just use the subdomain of primary product for cases like these?
Would even help in generating organic traffic there.
One benefit of using a new domain is that I can post on ProductHunt and Hackernews without getting blocked. However, I do link back to LampBuilder with a button on these mini apps.
Fundamentally yes, although these products are more inclined towards marketing per se (proxy leads to your core business), compared to actual revenue generating stream.
One of the aspect that one might consider is some sort of startup grant by hosting or ai services.
Correct, the goal is three-fold:
In my reflection post I share a bit more on how each of these is playing out as well.
any ball park figure of the cost behind these ones?
Assuming we did not get free credits, it would cost $50/month at this time. At the moment out of pocket is only $20/month + custom domain names. But technically I can get this to zero as well if I shift to AWS where we have credits.
I do run all the projects on the same instance which does save on time and costs.
Hi @topbridge, I wrote a longer post here: https://aisuperfounder.substack.com/p/3-mini-ai-apps-launched-in-16-days
I would guess, $0 zero on hosting (using Cloudflare pages) and marginal cost on OpenAi call, or maybe running on Google Cloud PaLM also on credits for startup or free tier.
Instead of promoting your apps here, it would be better if you wrote a case study on how you made these apps.
Thanks for the feedback! I'll definitely write up a case study to reflect on and share. Looking to cover the following:
Anything else you would like to get more info on?
please link your X account with the IH profile
Thanks for the heads up! I linked it now.
Hi @Shahidia, I wrote a longer post here: https://aisuperfounder.substack.com/p/3-mini-ai-apps-launched-in-16-days
Great POC's! What platform did you use to create them?
@bottiverstas I wrote a longer post here: https://aisuperfounder.substack.com/p/3-mini-ai-apps-launched-in-16-days
I am using a stack I am very familiar with which saves a lot of time. Django + Heroku + PostgreSQL + Redis + Cloudflare + OpenAI + Namecheap.
I created scripts to help me automate a lot of the initial code generation. Will share this in a follow-up post later this week.
Unlock the potential of WhatsApp Marketing on your own terms. Whether you're an entrepreneur eager to supercharge your business or aiming to start your own profitable SaaS venture.
Great, hopefully in the future we can share how to acquire users. These interfaces are really simple to the extent that users don't know how they will work or what they will bring them. Despite you bringing up some great ideas about the entrepreneurial whiteboard.
Thanks for the feedback! I am thinking about adding a simple "how it works" section on the page. Do you think that would help?
yoo this is really lovely! I love the consistent design style amongst the 3 of them
Hi @Drounz, I wrote a longer post here: https://aisuperfounder.substack.com/p/3-mini-ai-apps-launched-in-16-days
nice !!
Thank you! It helps to have time constraints - simplicity is a result thereof. Projects I spend more time on, tend to be over-engineered and end up being complex. A good exercise to build with constraints.
Wow and I thought I am doing good with my prototype timing. I am 2 weeks in on my project and still a few to go. 😅
Don't feel discouraged!
I spent a lot of prior time creating a base framework that takes care of tedious things like deploying, domain configurations, running tasks, etc.
On top of that, I also created a couple of scripts that generate a lot of the initial code. I'll share more about this later this week.
I appreciate your time allocation for a project, especially the restriction on development time.
Many indie hackers tend to fall into the trap of endless development and self-satisfaction, overlooking the importance of marketing.
It is so hard to get out of that trap. Personally, I am still in that trap. Looking to unpack this more in a reflection post.
Wow, that's really awesome to hear!
Thank you!
i can see all of them working as marketing channel of Lamp builder. Great.
Thanks, yes that is definitely a part of the plan here. It is interesting to see how it plays out as a marketing tactic. Maybe in the future, it could come together as a single project, but we are still trying to get a better idea of the use cases AI is good at. As a general area we are focused on the role of AI in starting/running startups.
Amazing! You really took an advantage of AI.
Thank you! So much to learn still :-)
i was blown with lean canvas ai. brought down the rabbit hole with office hour, and was lost at get feed back ai. will be waiting for some lesseons next week!
Thank you for testing it out!
Just tried all three, pleasantly surprised by the quality of feedback each provided. Not fully sure how it could be monetize. I like the create landing page integration as a start. Also, in each of these exercises I feel like part of the point of them is for us to think and evaluate. Having the AI do it is against the point (business plan evaluation, opportunity canvas creation).
I would be interested in how you made these too. Very good MVPs.
Thank you for testing it out! Still figuring out the monetization part. What is interesting is to see the different types of engagement/feedback for the respective projects. It does help to get early signals of what users are finding useful vs not.
a totally agree.
Congrats! Nice work launching. Curious what you ended up building with? Also curious how you prevent multiple submissions?
Hi @Esus, I wrote a longer post here: https://aisuperfounder.substack.com/p/3-mini-ai-apps-launched-in-16-days
Thanks for testing it out! I'll share a bit more about my stack in an upcoming post. Shared a short summary in a comment above.
In terms of multiple submissions. I don't have any rate limiting/spam protection at the moment. However, I am keeping an eye on things and do have a daily budget cap on Heroku/OpenAI credits. This helps with protecting against worse-case scenarios.